Non-alcoholic fatty liver disease (NAFLD) is a chronic liver condition characterized by insulin resistance and hepatic fat accumulation (steatosis). It affects ~80 million Americans (Browning et al, 2004). In ~30-40% of patients steatosis leads severe steatohepatitis with necroinflammation and fibrosis (i.e., nonalcoholic steatohepatitis or NASH). NASH may progress to cirrhosis in ~10% of patients. Lifespan is not only shortened by liver-related morbidity, but also by a higher incidence of cardiovascular disease (Targher et al, 2007). My long-term goal is to contribute to our understanding of the host of factors that cause NASH in Hispanics with T2DM and improve the current diagnostic and treatment strategies. The objectives of this application, which are the next step in pursuit of that goal, are to determine the true magnitude of the NAFLD epidemic in Hispanics with T2DM in South Texas by means of a systematic screening approach (i.e, magnetic resonance spectroscopy or MRS) (Aim #1), and to assess the long-term safety and efficacy of pioglitazone (Aim #2). In the first controlled trial of its kind, we have reported that 6 months of pioglitazone reverse metabolic/histological defects in NASH (Belfort et al, NEJM 2006). The central hypothesis of this application is that that the true prevalence of NAFLD is much higher than previously appreciated and that long-term treatment with pioglitazone alters the natural history of the disease. In Aim 1 subjects undergo an initial history/physical and routine labs, to exclude other conditions known to cause a fatty liver. Liver fat (by MRS) is then measured in eligible patients. We will also assess: i) insulin secretion by an OGTT, ii) muscle/live/adipose tissue insulin resistance by a euglycemic insulin clamp; iii) total body fat by DXA, to assess the relationship between adiposity and liver fat; iv) a liver biopsy if they have a fatty liver (MRS) and a high-risk profile for NASH. Subjects with biopsy-proven NASH will be invited to Aim #2, where 120 patients will be randomized (by a computer-based algorithm) to dietary intervention plus pioglitazone (45 mg/d) vs. dietary intervention (the current standard of care) plus placebo. Glycemic control will be optimized following ADA guidelines. Patients will be followed q1-2 months and liver MRS repeated every 6 months. The metabolic studies (as in Aim#1) and liver biopsy will be repeated at 18 months. This study will define for the first time the prevalence of NAFLD in Hispanics with T2DM. The clinical, metabolic and molecular (liver biopsy) evaluations will identify variables associated with disease progression and help understand how pioglitazone works. Without this knowledge, we cannot move beyond lifestyle recommendations to treat NASH. There are no pharmacological treatments. NASH is an increasing problem across VAs. Moreover, obese and diabetic Hispanic patients have the highest risk of disease progression. Therefore, this project is important to our San Antonio VA (and the entire VISN 17) where the prevalence of obesity is ~70%, that of diabetes ~30 % and more than 50% of patients are of Hispanic ancestry.